Many cities have recently developed and passed benchmarking laws (also known as energy disclosure, information, or transparency laws) to make energy information available for individual buildings. This building-level micro-data can be used to better design and develop municipal policies for energy efficiency. Based on benchmarking data collected by the City of New York, this paper presents a comprehensive portrait of energy use in New York City commercial and multifamily buildings, including the distribution and concentration of energy uses and building types within the overall population. First, this paper describes energy and building characteristics for the overall population of over 10,000 buildings, comprising over 1.5 billion square feet. Second, using model-based clustering methods, this paper then identifies key clusters of energy use and building characteristics in the multifamily sector. The paper finds that these clustering methods describe sub-groups in the population in intuitive ways. Third, using multivariate regression, the identified clusters are then used to improve predictions of building energy use and the targeting of prospective efficiency efforts. These methods and results should be of interest to energy and policy researchers in New York and other cities.
[1]
H. Bock.
Probabilistic models in cluster analysis
,
1996
.
[2]
L. Schipper,et al.
Overcoming social and institutional barriers to energy conservation
,
1980
.
[3]
Adrian E. Raftery,et al.
Model-Based Clustering, Discriminant Analysis, and Density Estimation
,
2002
.
[4]
A. Henningsen,et al.
systemfit: A Package for Estimating Systems of Simultaneous Equations in R
,
2007
.
[5]
Jeffrey M. Woodbridge.
Econometric Analysis of Cross Section and Panel Data
,
2002
.
[6]
H. Allcott,et al.
Is There an Energy Efficiency Gap?
,
2012
.
[7]
A. Zellner.
An Efficient Method of Estimating Seemingly Unrelated Regressions and Tests for Aggregation Bias
,
1962
.
[8]
Adrian E. Raftery,et al.
MCLUST Version 3 for R: Normal Mixture Modeling and Model-Based Clustering †
,
2007
.
[9]
A. Raftery,et al.
Model-based Gaussian and non-Gaussian clustering
,
1993
.